9 research projects participated this year in a 3 days’ intense workshop for academics who want to challenge the potential to transfer their technology/their research results into successful products or services/ potential spin-off.
Fecally contaminated drinking water may harbor many different pathogenic microorganisms and therefore poses a major threat, especially to people in developing countries and in crisis regions. Current methods for the detection of fecal indicator bacteria in water take more than 18 hours or involve costly instruments and high logistical effort. Our project comprises a novel approach that enables such analyzes in under 1 hour with minimal equipment directly at the point of interest. Thus, the assessment of microbiological water quality is made accessible in credit-card format also to people in remote areas without large financial resources or analytical background. In further consequence, the resulting prototype will serve as a model for a large number of future molecular diagnostic solutions, for example in the food or animal feed sector.
The Artificial Researcher – Science (AR-Science) automatically retrieves the publications based upon the user request, and narrow down to the specific text segments that are defined by the users' information needs. AR-Science also provides users with the background research in terms of argumentative zoning, and comparative documentation summarisation, and thereby reduce the time spent on searching and reviewing related work. By providing the users with a unified platform , which supports the complete research activity, we reduce the time-consumption for each user concerning search and reviewing scientific publications. At the same time, we provide libraries and other information providers with essential information for future support of research activities and a purchase strategy for obtaining scientific literature.
The Bee Vision system (camera box and mobile app) will recognise the biggest threat to the honeybee, the parasitic mite Varroa destructor, in real time without affecting the bees.
Based on innovative methods of digital image processing and machine learning implemented on a single board processor, an application with a camera system will be built to attach to the entrance of the hive.
Bees with mites will be detected in real time and Bee Vision will be able to reliably quantify the infestation and give automatic, precise and remote warning (mobile app) at the optimal time to react, as current counting methods for the Varroa infestation of a hive are invasive, non-remote, difficult, time-consuming and neither precise nor reliable.
With additional features of the camera system, e.g. precise bee counting, the health of a colony can be monitored to avoid colony losses.
The blockchain technology, especially Bitcoin (BTC) has received much attention during the past 2 years. Up to now in most of this research machine learning has been excluded. But particular attention should be paid to analyzing the blockchain with neural networks (NN), i.e.
Recurrent NN (RNN) or Convolutional NN (ConvNet). The reason for this is that RNNs provide the ability for processing sequences and Bitcoin transactions are fully linked where money is transferred. This is one of the base cases in analyzing the blockchain.
Keep Distance - Slow Down, Danger Ahead!
Keepdistance offers a simple solution to increase the proximity road safety level, based on a real-time community-shared information. Once the app is correctly installed on the smartphone of a user, Keepdistance automatically filters outgoing emergency calls from the device, it geo-localizes the device, and it shares in real time the information with any users in motion over a certain speed in a given radius from the device through a push notification sent to the users' smartphone. The greatest advantage of Keepdistance stems from the passive interaction of the user making the call with the platform Keepdistance.
Sustainable Membrane Technology for World Water Scarcity
As many parts of the world are reaching a critical situation of water scarcity causing the next great global crisis reported by WHOi, a newly evolved membrane distillation (MD) technology provides a promising opportunity for drinkable water production even in remote and rural areas shortcoming water treatment systems. This proposed modified desalination process can be coupled to low-grade and renewable energy sources such as solar energy to remove the salt from seawater and brackish water. With the availability of specifically designed membranes, this proposed technology is at the threshold of commercialization.
At Simple-AI we develop new machine learning technologies for real-life applications such as smart predictive maintenance platforms for industrial robots as well as brains for self-driving cars.
The organ-on-a-chip technology was described by the World Economic Forum in 2016 as one of Top 10 "emerging technologies". To meet the growing need to improve the predictive power of toxicological, pharmaceutical and preclinical studies, the MultiSphere project builds on the advancement of a novel organ-on-a-chip technology that establishes, cultivates and supports a wide variety of multicellular spheroids (3D cell aggregates). The MultiSphere project thus closes an important technological gap, enabling rapid and easy production of spheroids of defined cell types. With the help of our biochip, each cell type (primary, cancer or stem cell) can now be established, treated and analyzed as a high-throughput in vitro human organ model. Since the present spheroid biochip technology is compatible with standard cell biology measurement techniques (e.g., microscopy, spectroscopy, etc.), the MultiSphere platform enables cost-effective and time-saving monitoring of 3D cell cultures in a standardized and reproducible manner.
We want to develop innovative concepts for aircrafts to counteract effects of in-flight turbulence. By means of a new approach for flight control, which makes use of highly dynamic wing deformation, adverse acceleration effects are drastically reduced. Turbulence prediction, which combines sensor and turbulence modell data, further enhances the flight control allowing for proactive countermeasures. The overall solution is targeted to provide independence of current weather conditions. This improves availability, predictability, and reliability for flight operators - from UAV to airliner - as well as passengers and costumers.
Over the course of the 3 days, the participating teams worked on defining a personalized framework for rigorously planning their market entry and validating market assumption with the help of 36 experienced national & international mentors, coaches, investors and industry representatives. Through a combination of input sessions, mentoring sessions and peer-review sessions, the potential spin-offs got to identify the commercial aspects of their research projects and learned how to present and communicate their ideas in a short, comprehensible and compelling manner to non-experts.
On the 4th day, the teams got the chance to present the outcome of the camp during a Pitch-event with a jury of experts at the i²c Networking Friday for the chance to win various prizes.